Topics

No smoking guns under the Sun

30 October 2000

A range of different experiments have studied in detail the neutrinos emitted by the Sun. What does this complex picture now tell us? Arnon Dar reviews the latest wisdom.

cernsun1_11-00
cernsun2_11-00

The Sun is a typical main sequence star that generates its energy via the fusion of hydrogen into helium in two chains of nuclear reactions: the so-called pp chain and the CNO chain. If the nucleon number, electric charge, lepton flavour and energy are conserved and the Sun is in a steady state, then the total solar neutrino flux is fixed, to a good approximation, by the solar luminosity (approximately 65 billion neutrinos/cm2/s at Earth), independent of the specific nuclear reactions that power the Sun and produce neutrinos by beta decay or the electron capture of reaction products.

The neutrinos from the dominant pp chain are produced by the beta decay of proton pairs (pp), boron-8 and lithium-4, and by electron capture by pp pairs and beryllium-7. Their spectra can be measured directly in the laboratory or calculated from the standard theory of electroweak interactions.

To a very good approximation, they are independent of the conditions in the Sun. Only their relative contributions depend on the detailed chemical composition, temperature and density distributions in the Sun. Solar neutrino experiments can therefore test both the standard theory of stellar evolution and neutrino properties over a long distance, much larger than the diameter of Earth.

By the turn of the last century, solar neutrinos had been detected by radiochemical methods in three underground solar neutrino experiments in the US (Homestake) and Europe (SAGE and GALLEX) and in real time by the water Cherenkov techniques in two experiments in Japan (Kamiokande and Superkamiokande). These studies have confirmed that the Sun is powered by the fusion of hydrogen into helium – a milestone achievement in physics.

However, the combined results also suggested that the solar neutrino fluxes differ significantly from that expected from the standard solar models. This discrepancy has become known as the solar neutrino problem (SNP).

Neutrino oscillations

Many scientists have argued that this discrepancy is due to neutrino properties beyond the minimal standard electroweak model. In 1968, Gribov and Pontecorvo suggested that “oscillations” of electron neutrinos to other neutrino flavours may reduce the solar electron neutrino flux at the Earth. Later, Mikheyev and Smirnov elaborated on work by Wolfenstein on the propagation of neutrinos in matter and found that matter amplification of these oscillations in the Sun can provide an elegant solution (the so-called MSW solution) to the SNP.

The widespread belief in this solution of the SNP was strengthened by the accumulating data from the deep underground experiments on the atmospheric neutrino anomaly (fewer muons than expected) and, most recently, also from the first terrestrial long distance neutrino experiment, K2K, which were reported by Kenzo Nakamura from the KEK laboratory at Neutrino 2000 – the 19th international conference on neutrino physics and astrophysics which was held this summer in Sudbury, Canada.

Both the atmospheric neutrino anomaly and the K2K results can be explained by the hypothesis of nearly maximal strength oscillations of muon neutrinos to tau neutrinos if their squared masses differ by some 3 x 10-3 eV2. However, conclusive solar neutrino evidence for electron neutrino properties beyond the standard electroweak model can be provided only by detecting at least one of the following signals:

* neutrinos other than electron-type visible by neutral current interactions;

* spectral distortion of the fundamental beta-decay spectra;

* a neutrino flux different from that expected from the solar luminosity;

* modulations of the solar neutrino flux, such as a day-night or summer-winter difference, other than that expected from the seasonal variation in the Earth’s distance from the Sun.

Looking for smoking guns

It was hoped that these “smoking gun signals” would be found before Neutrino 2000 with the two currently operating solar neutrino telescopes: the 50 kt Superkamiokande underground light-water Cherenkov detector that has been collecting data on the boron-8 solar neutrino flux, its spectrum, and seasonal and day-night possible variations, with a lower energy threshold and lower background; and the 1 kt Sudbury Neutrino Observatory (SNO) heavy-water detector in a 2 km deep Canadian mine that started taking data half a year ago and is expected to detect the conversion of solar electron neutrinos to mu or tau neutrinos through their dissociation of the deuterium into a proton and a neutron in the heavy water.

However, no such signals have been detected. At Neutrino 2000, Yoichiro Suzuki from the Kamioka Observatory presented data from 1117 days running of Superkamiokande which show no day-night effect, no spectral distortion of the boron-8 solar neutrino spectrum, and the expected variation due to the annual variation in the distance between the Earth and the Sun.

In fact, the use of a new and more precise laboratory measurement of the neutrino spectrum from boron-8 beta decay and a new estimate of the cross-section for proton capture on helium-3 yield an excellent agreement between the expected and observed Superkamiokande solar neutrino spectra as seen in figure 1.

When combined with the other solar neutrino experiments, the Superkamiokande data rule out, with 95% confidence, the small mixing angle MSW solution and a “sterile” neutrino oscillation solution to the SNP. It leaves only a small region in the mass-mixing exclusion plot with a large mixing angle as a possible simple oscillation solution to the SNP as can be seen from figure 2. Fortunately, this solution will be tested in the near future in a terrestrial experiment – KamLAND, a long base-line neutrino oscillations experiment in Kamioka using nuclear reactor neutrinos, and with new solar neutrino experiments such as BOREXINO.

What if?

But what if the large mixing angle oscillation solution to the SNP will also be ruled out by KamLAND and BOREXINO, and SNO will not detect conversion of solar electron neutrinos into mu or tau neutrinos? At Neutrino 2000, E Belloti, spokesman of the Gallium Neutrino Observatory (GNO), and V Gavrin, spokesman of SAGE, reported updated results for the solar neutrino capture rate in gallium. Their measured rates, some 78 ± 7 standard solar neutrino units (SNU), are above the minimal signal expected from the observed solar luminosity, if solar neutrinos do not oscillate.

These results seem to leave only a little room for solar neutrinos from electron capture by beryllium-7 in the Sun. This is also suggested by the results from the pioneering chlorine experiment of Ray Davis at Homestake, the counting rate of which, 2.56 ± 0.23 SNU, is consistent with the solar neutrino flux (2.37 x 1010 cm2/s) measured by Superkamiokande, but leaves very little room for beryllium-7 neutrinos.

However, this conclusion heavily relies on the accuracy of the theoretically-deduced cross-sections for neutrino capture in gallium and chlorine. If the results of GALLEX and SAGE are calibrated with their chromium source experiments, they leave more space for beryllium-7 solar neutrinos, perhaps sufficient to accommodate a solar electron-capture rate in beryllium-7 consistent with the solar proton-capture rate in beryllium-7 that produces the observed boron-8 solar neutrino flux in Superkamiokande.

A direct calibration experiment for the chlorine detector was described by Ken Lande at Neutrino 2000. Improved calibration experiments are also under consideration by the GNO and SAGE collaborations. Altogether, it will still require long, challenging and innovative experiments to give a complete spectroscopy of the elusive solar neutrinos and pin down the origin of the SNP.

bright-rec iop pub iop-science physcis connect